Which factors are associated with cannabis use among adolescents in Andalusia? An application of the I-Change model

J Nurs Sch. 2023;55:739–750. | 739wileyonlinelibrary.com/journal/jnu Received: 24 March 2022 | Revised: 21 July 2022 | Accepted: 24 July 2022 DOI: 10.1111/jnu.12806 SPECIAL ISSUE Which factors are associated with cannabis use among adolescents in Andalusia? An application of the IChange model MaríaCarmen TorrejónGuirado MHS1,2 | Marta LimaSerrano PhD1 | Liesbeth Mercken PhD2,3 | Hein de Vries PhD2 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. Journal of Nursing Scholarship published by Wiley Periodicals LLC on behalf of Sigma Theta Tau International. 1Department of Nursing, School of Nursing, Physiotherapy and Podiatry, University of Seville, Seville, Spain 2Department of Health Promotion, Care and Public Health Research Institute CAPHRI, Maastricht University, Maastricht, The Netherlands 3Department of Health Psychology, Open University Heerlen, Heerlen, The Netherlands Correspondence Marta LimaSerrano, PhD, Department of Nursing, School of Nursing, Physiotherapy and Podiatry, University of Seville, Seville, Spain. Email: [email protected] Funding information MCIN/AEI, Grant/Award Number: PID2019107229RAI00; VPPIUniversidad de Sevilla Abstract Background: Cannabis use is increasing among Andalusian adolescents, causing health, social and financial problems. School nurses visit schools and provide health promotion to adolescents in Andalusia. In order to better tailor health promotion programs, it is important to understand the sociodemographic and motivational factors related to cannabis use in Andalusian adolescents. Design/Methods: In this crosssectional study, 369 students (aged 15– 18 years) from secondary schools in Andalusia were involved. An anonymous questionnaire based on the IChange Model was selfadministered during school hours. Cannabis users were compared with noncannabis users regarding sociodemographic and motivational factors. Contingency tables, mean comparison tests, and logistic regression analyses were conducted. Results: The prevalence of cannabis use in the last year was 21.6%. Noncannabis users had a few positive beliefs toward cannabis use (e.g. the ability to relax or medical benefits), as well as users. Users recognized some disadvantages of cannabis use but played down their importance and mentioned more advantages. Moreover, social influence, especially from peers, plays a critical role in cannabis use. Specific social situations and moods seemed to be special risk factors for cannabis use, such as being at a celebration or feeling good. Results of regression analysis showed that cannabis use is mainly associated with age, low family function, low family affluence, high pocket money, perceiving few disadvantages of cannabis use, much social modeling of cannabis use, social norm and pressure favoring cannabis, low selfefficacy to resist using cannabis and positive intentions to use cannabis. Conclusions: Based on our results, prevention programmes lead by nurses can be tailored to the factors important to prevent cannabis use, focusing on outlining the disadvantages of cannabis use, alternatives for using cannabis when feeling good, increasing salience of healthy social influences and reinforcing selfefficacy to resist temptations to use cannabis are recommended. 740 | CANNABIS USE FACTORS IN ADOLESCENTS BACKGROUND Cannabis is the third most used drug worldwide by adolescents, with considerable implications for public health (European Monitoring Centre for Drugs and Drug Addiction, 2021; Spanish Observatory of Drugs and Addictions, 2021; World Drug Report, 2021). It is mainly used mixed with tobacco, and cannabis use is a risk factor for noncommunicable diseases, such as cardiovascular or pulmonary diseases, and mental problems, such as schizophrenia, psychosis and bipolar disorder (European Monitoring Centre for Drugs and Drug Addiction, 2021; Spanish Observatory of Drugs and Addictions, 2021; Spanish National Plan on Drugs, 2018; World Drug Report, 2021). The early exposure of adolescents to cannabis may cause important changes in the reactivity of the brain, which makes adolescents more vulnerable than other age groups (Spanish National Plan on Drugs, 2018). In addition, epidemiological studies analyzing the effects of an early onset of cannabis use have reported increased risks for developing drug and alcohol use disorders and an increased risk for mental health problems (Blanco et al., 2016). Moreover, cannabis use is associated with an increased likelihood of progression to hard drugs and accounts for 26% of European emergency visits for intoxications (European Monitoring Centre for Drugs and Drug Addiction, 2021). Despite these potential consequences, the levels of perceived risk and disapproval of cannabis among adolescents are low and steadily declining health (European Monitoring Centre for Drugs and Drug Addiction, 2021; World Drug Report, 2021). Cannabis use in adolescents has increased at a global level (World Drug Report, 2021). In Europe, each country has its own cannabis regulation rules, however its use is only legal in the Netherlands. Spain, and specifically Andalusia due to its proximity to Morocco, has been identified as the main entrance and transit country for cannabis in Europe (European Monitoring Centre for Drugs and Drug Addiction, 2021; Spanish Observatory of Drugs and Addictions, 2021), where cannabis is positioned as the most accessible illegal substance (Spanish National Plan on Drugs, 2018). These data endanger the biopsychosocial health of Andalusian adolescents due to the ease of access to cannabis in the region where they live. Furthermore, 25.5% of Spanish deaths in 2019 were associated with cannabis use. Figures from the Spanish Observatory on Drugs and Addictions (2021) have demonstrated an upward trend of individuals who begin to consume cannabis. In 2021, the Spanish ESTUDES study with a sample size of 22,321 students aged 14– 18 years has revealed that 4955 (22.2%) of these students used cannabis in the last 12 months. This data is higher than the average reported among European adolescents aged 15– 16 years (17.3%). The average age of first use in Spain is 15 years (Spanish National Plan on Drugs, 2018). Moreover, 29% (n = 6473) of Spanish adolescents aged 14– 18 years have used cannabis at least once in their life and 14.9% (n = 3326) in the last month. Andalusia, the most southern community of Spain with a population of 8.4 million inhabitants, is encountered with the same trend, showing a progressive increase in cannabis use over the years, that is, 1.4% in the last 6 months and 0.4% in the last month. Cannabis use in Andalusia is prevalent resulting in a reported use by 36.7% (n = 260) of the boys and 32.3%(n = 112) of the girls of 15– 17 years old for a sample size of 1055 participants (Andalusian Government, 2018). Moreover, 36% (n = 94) of a sample of 261 Andalusian adolescents aged 16– 20 years have used cannabis at least once in their life, 23.8% (n = 62) in the last year and 15.7% (n = 41) in the last month (Agency for Social Services and Dependency of Andalusia, 2018). On the other hand, results of the Health Behavior in Schoolaged Children Study (HBSC2018) in Andalusia with a sample size of 1334 adolescents between 11– 18 years old shows that 22% (n = 294) have ever tried cannabis at least 1 time in their life (Moreno et al., 2020). As cannabis is an illegal substance in Andalusia, these data may be subject to underreporting. The data on adolescents' admissions in Andalusian public center for drugs treatment show that 727 minors of a total of 19,946 individuals were admitted for drug treatment in Andalusia. Of these 727 adolescents, 621 (85.4%) were admitted for cannabis dependence treatment (Andalusian Government, 2018). These data indicate that admission to drug treatment in Andalusian adolescents is higher for cannabis use than for other substance, such as alcohol and tobacco. Moreover, studies on cannabis data in Andalusia adolescents are very scarce, and suffer from low sample sizes and outdated data. To our knowledge, no previous Andalusian study has examined the prevalence of cannabis use in Andalusian adolescent nor their motives for using it. Looking at these data, although some Spanish studies have examined the predictors of cannabis use, more evidence using Andalusians as a target population is required (Becoña et al., 2013; LiébanaPresa et al., 2020; Lloret Irles et al., 2018; Morales et al., 2008; Rial et al., 2018). For developing a tailored approach to prevent cannabis use among Andalusian adolescents, the main sociodemographic and motivational factors underlying cannabis use must be identified. To understand the motives for cannabis use, several models such as the Precaution Adoption Model (Sharma, 2007), the Health Belief Model (Janz & Becker, 1984), Theory of Planned Behavior Clinical Relevance: Nurses need to know the sociodemographic factors and motivations associated with the use of cannabis in adolescents in order to establish effective preventive interventions at school. KEYWORDS adolescents, cannabis, health behavior, IChange model, motivation 15475069, 2023, 3, Downloaded from https://sigmapubs.onlinelibrary.wiley.com/doi/10.1111/jnu.12806 by Universidad De Sevilla, Wiley Online Library on [06/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 741 CANNABIS USE FACTORS IN ADOLESCENTS (Ajzen, 1991) and goal setting theories (Locke & Latham, 1990) can be used. One framework that integrates findings from several of these cited cognitive models is the I Change model (ICM) (De Vries & Mudde, 1998). We selected this model for our study given its suitability and comprehensiveness as well as it has already been used to explain the consumption of substances such as alcohol and tobacco (MartinezMontilla et al., 2020; Smit et al., 2018). The ICM recognizes three phases in the process of behavior change: awareness (determined by factors, such as cognizance, knowledge, risk perception and cues to action), motivation/intention (determined by attitude, social influence beliefs and selfefficacy) and action (determined by action planning, preparatory planning, coping planning and plan enactment). The motivational phase resembles the constructs of the Theory of Planned Behavior but slightly differs in their operationalisation and addition of social modeling to social norms (De Vries et al., 1995). Several studies using socio– cognitive models such as the Theory of Planned Behavior (De Vries et al., 1995; Frank & Fiegel, 2020; Jalilian et al., 2020) and Social Cognitive Theory (De Vries et al., 1995) have shown the importance of attitudes, social norms, social modeling, selfefficacy and intentions. An important elaboration of the ICM is the acknowledgement of factors that determine a person's awareness, motivation and actions, such as family and social factors. For cannabis use, studies (Guxens et al., 2007; Terzic Supic et al., 2013; Trujillo et al., 2016) have revealed that adolescents may be at an increased risk of using cannabis when they are living in dysfunctional families, when they have an impaired relationship with parents, or when they have little involvement with parents. However, concerning the influence of socioeconomic status, studies have revealed mixed findings. Some (Legleye et al., 2012; Rogeberg, 2013) have suggested that low socioeconomic status is related to increased cannabis use, whereas others (Guxens et al., 2007; McCrystal & Percy, 2019) have reported more cannabis use among adolescents who have more pocket money to spend. Nurses can play an important role on cannabis prevention on adolescents. They are trained to value people as biopsychosocial beings. This professional competence is highly related with the success of a behavior change (Mayfield & Fogger, 2022). Nurses have good competences about health promotion, they are trained in emotional intelligence and social skills, and they have access to schools and adolescents. Interventions offered by nurses are becoming more individualized and comprehensive (Halladay et al., 2020). Therefore, school nurse can easily establish a bond of trust with the adolescent. This characteristic would facilitate the success and adherence of cannabis prevention interventions. Despite the advantages of nursing leadership in cannabis prevention, there are very few studies on the nursing implications of cannabis prevention in adolescents (Halladay et al., 2020; Mayfield & Fogger, 2022). Studies targeting adolescents sociodemographic and motivational factors in Spain and Andalusia are relatively scarce. Therefore, this study was designed to identify the role of sociodemographic and motivational factors related to cannabis use in the last 12 months among adolescents aged 15– 18 years, as well as family and social factors, to identify variables to be addressed when developing a cannabis prevention programme lead by nurses. METHODS Research design and sample In 2018, crosssectional study was conducted, in the fourth grade of compulsory secondary education (CSE; equivalent to the 10th grade in the USA), the first baccalaureate (equivalent to the 11th grade in the USA) and vocational training (VT). Three secondary schools from the provinces of Sevilla, Cadiz and Huelva (western Andalusia) agreed to participate; six classes from each school participated in this study. Power calculations suggested a sample size of 308 participants for logistic regression using G*Power 3.1.9.4, for an α error of 5%, a power of 80% and an odds ratio of 1.5 (twotailed test). Finally, 369 students participated in this study. Variables and instrument A selfadministered questionnaire with the following variables was used. Sociodemographic variables Gender (0 = male and 1 = female), age (in years), nationality (dichotomised into 1 = Spanish and 2 = other), religion (1 = Catholic, 2 = others and 0 = no religion), adolescent's educational level (1 = CSE, 2 = baccalaureate and 3 = VT) and parents' educational level (0 = none, 1 = CSE, 2 = baccalaureate and 3 = university studies) and having (1) or not having (0) a boyfriend/girlfriend. Social status was measured using the Family Affluence Scale (FAS) (Cronbach's α = 0.26), which comprises four questions for assessing whether students had a low (0– 2 points), moderate (3– 5), or high social status (6– 9) (Hartley et al., 2016). Family functioning was assessed using the Family Apgar Test (Cronbach's α = 0.77), consisting of five questions, answered on a 3point Likert scale (from 0 = almost never to 2 = almost always) (Smilkstein et al., 1982). Finally, weekly pocket money assessed the amount of money an adolescent can spend (0 = €0; 1 = up to €10; 2 = €11– €20; 3 = €21– €30; 4 = >€30; 5 = on demand, that is, the adolescent has money every time they request it; the student does not receive a fixed amount). Cannabis use was measured using the frequency of use in the last 12 months (0 = no cannabis use; 1 = cannabis use in the last 12 months) (Spanish National Plan on Drugs, 2018). 15475069, 2023, 3, Downloaded from https://sigmapubs.onlinelibrary.wiley.com/doi/10.1111/jnu.12806 by Universidad De Sevilla, Wiley Online Library on [06/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 742 | CANNABIS USE FACTORS IN ADOLESCENTS Motivational factors For details on the various items, see Table 1. Attitude toward cannabis was assessed using a 5point Likert scale (1 = strongly disagree; 5 = strongly agree) with six items assessing the advantages and six items assessing the disadvantages. For the regression analysis, the items were summed into a pro scale (Cronbach's α = 0.83) and a con scale (Cronbach's α = 0.86). Social norms were assessed using the perceived opinion of the social environment on whether the respondent could consume cannabis or not (8 items; 5 = definitely, it is okay to use cannabis; 1 = definitely, it is not okay to use cannabis) (Cronbach's α = 0.89). Social modeling assessed the frequency of cannabis consumption by individuals from the social environment of the respondent using a 5point Likert scale with 10 items (0 = never; 4 = always; 8 = I do not have) (Cronbach's α = 0.75). Social pressure assessed the extent the respondent has encountered pressure to use cannabis by individuals from their social environment using a 5point Likert scale with 10 items (0 = never;4 = always; 8 = I do not have) (Cronbach's α = 0.93). Selfefficacy assessed the perceived difficulty of adolescents to avoid cannabis use in different situations using a 5point Likert scale with 11 questions (0 = very difficult; 4 = very easy) (Cronbach's α = 0.94). Intention of cannabis use in the next year and in the future was assessed using a 5point Likert scale (0 = surely no; 4 = surely yes) (Cronbach's α = 0.91). Procedures The heads of the schools that participated were contacted via telephone to request their approval and collaboration, and when approved, they were followed through contacts with school counselors to arrange the visits. Information about this study, as well as informed consent and questionnaires, was sent to the school heads, counselors and parents to decide their participation. Passive written informed consent was requested from the parents and active consent from the participants. The questionnaires were administered in school classes and lasted approximately 30 min, supported by two researchers. The students voluntarily responded in a paper format. The confidentiality of the data was explained to the students, and the procedures followed the Regulation (EU) 2016/679 of the European Parliament and the Data Protection Council. The study was approved by the Research Ethics Committee of Andalusia (registration number: 0073N18). Data analysis To identify the characteristics of the sample, descriptive analyses were conducted. First, we examine every item separately. Then, taking as a cutoff point of 0.60 for Cronbach's alpha, scales were computed for attitude, social influences, selfefficacy and intention, which were used in the regression analysis. To analyze the differences between the two groups (users and nonusers), the chisquare test was used for categorical variables and ANOVA for numerical variables. Cannabis use in the last 12 months was the dependent variable. To assess how well the measured factors are associated with cannabis use, binary logistic regression analysis was conducted (Field, 2017). To detect suppressor effects, Pearson's correlations were analyzed (Table 2). All variables were included in the model, except for the city of residence, parents' educational level, having or not boyfriend/ girlfriend, religion and nationality, because they have pvalues of more than 0.05 in the chisquare test. To test the model's fit, variables were entered by blocks of predisposing factors. “On demand” of pocket money and “I do not have that relative” in social modeling and pressure were coded as missing values due to their difficult interpretation. Furthermore, Apgar and FAS were used as indices because of their multidimensional nature. Model 1 contained sociodemographic variables (i.e. age, gender, Apgar, FAS and pocket money), model 2 consisted of motivational factors (i.e. advantages, disadvantages, social influence and selfefficacy) and model 3 contained the intention to cannabis use. The level of significance was set at p < 0.05, and the enter method was used in this study. Although there were very few missing values (see Table 3), a multiple data imputation was performed. This statistical analysis was conducted using Statistical Package for the Social Sciences (version 26; IBM Corporation, 2019). RESULTS Characteristics of the sample by cannabis use in the last 12 months Table 3 shows the characteristics of the overall sample and the differences between users and nonusers. In this study, 21.6% of the students have tried cannabis in the last 12 months. The mean age of the respondents was 16.42 years. Most respondents were girls (53.4%) and in the 10th grade (56.1%). Most students had parents with a 10thgrade education level and did not have a boyfriend/ girlfriend (62.6%). Most students were of Spanish origin (98.4%) and were Catholic (69.1%). Users were significantly older and more likely to be male and belong to dysfunctional families and families with a lower affluence than nonusers. Concerning pocket money, Table 3 suggests that cannabis users more often receive pocket money of €20, whereas nonusers receive pocket money on demand. Attitude toward cannabis use All participants were convinced of the disadvantages of using cannabis (Table 1); however, users were significantly less convinced than nonusers of the disadvantages of using cannabis, such as feelings of regret, bad health and family problems. Although all participants perceived several advantages of using cannabis, users were 15475069, 2023, 3, Downloaded from https://sigmapubs.onlinelibrary.wiley.com/doi/10.1111/jnu.12806 by Universidad De Sevilla, Wiley Online Library on [06/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 743 CANNABIS USE FACTORS IN ADOLESCENTS TABLE 1 Mean and F ANOVA scores of beliefs, social modeling, normative beliefs and social pressure and selfefficacy and intention; comparison of users with nonusers General mean Mean nonusers Mean users F pN = 364 N = 294 N = 70 Disadvantages (m 5)a If I (were to) use cannabis: … this is bad for my health 4.24 4.47 3.26 98.744 0.000 …gives me family problems 4.15 4.38 3.21 90.250 0.000 … gives me conflicts with peers 3.69 3.94 2.66 69.364 0.000 … I do not like how I am 3.14 3.37 2.19 58.044 0.000 … makes me feeling guilty 3.45 3.65 2.57 51.191 0.000 … generates feelings of regret if I get sick after using 3.95 4.11 3.29 33.270 0.000 Advantages (m 5)a If I (were to) use cannabis: … it makes me feel relaxed 3.59 3.40 4.37 35.472 0.000 …can be used as a medicine 3.67 3.51 4.31 21.922 0.000 … stimulates my creativity and imagination 2.58 2.44 3.19 17.434 0.000 … makes me feel more secure with my friends 2.65 2.54 3.11 13.742 0.000 … makes me have fun 2.58 2.47 3.07 12.450 0.000 … helps to forget problems 2.84 2.73 3.27 8.064 0.005 Social norm (m 10)b Who thinks that you can use cannabis? My mother thinks … 1.22 1.21 1.26 0.100 0.752 My father thinks … 1.39 1.35 1.56 1.310 0.253 My sister (s) thinks (n) … 2.97 2.96 3.01 0.021 0.884 My brother (s) thinks (n) … 2.83 2.79 2.99 0.260 0.611 My Friends (male) think … 1.70 1.51 2.49 43.772 0.000 My friends (female) think … 1.65 1.47 2.49 57.414 0.000 My best friend thinks … 1.73 1.57 2.40 20.639 0.000 My boyfriend/girlfriend thinks … 3.05 3.07 2.96 0.088 0.767 Social modeling (m 5)c Who of the people mentioned use cannabis…? … my mother 0.10 0.11 0.09 0.052 0.820 … my father 0.40 0.32 0.73 4.102 0.044 …my sisters 2.32 2.38 2.09 0.381 0.538 …my brothers 1.97 1.97 1.96 0.001 0.972 … my best friend 0.66 0.48 1.67 25.193 0.000 …my friends (male) 0.71 0.41 1.53 72.533 0.000 …my friends (female) 0.62 2.24 2.46 75.022 0.000 … my classmates (male) 0.38 0.49 0.74 8.385 0.004 … my classmates (female) 0.54 0.32 0.60 5.637 0.018 …my boyfriend/girlfriend 0.63 2.24 2.46 75.022 0.000 Social pressure (m 5)c I have felt pressured to use cannabis for… My mother 0.12 0.13 0.09 0.198 0.656 (Continues) 15475069, 2023, 3, Downloaded from https://sigmapubs.onlinelibrary.wiley.com/doi/10.1111/jnu.12806 by Universidad De Sevilla, Wiley Online Library on [06/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 744 | CANNABIS USE FACTORS IN ADOLESCENTS significantly more convinced than nonusers of some advantages of cannabis, such as feeling relaxed and medical benefits. Social influence toward cannabis use Overall, regarding social norms, the respondents' social environment was negative about cannabis use. Most positive norms about cannabis use for both groups were reported by the boyfriend/girlfriend and siblings. Cannabis users reported a significantly (p < 0.05) more favorable social norm toward cannabis use from their best friend and male friends than nonusers. Concerning social modeling, the respondents mostly reported that few individuals in their social environment have used cannabis. Cannabis was most used by siblings in both groups. Nevertheless, users more frequently reported that cannabis was used by their best friends and male friends than nonusers. The participants reported relatively low levels of pressure to use cannabis. Both groups reported the highest levels of pressure to use cannabis from siblings and their boyfriend/girlfriend. Users felt significantly more pressure from their male friends, best friends and classmates than nonusers (Table 1). Selfefficacy toward cannabis use Cannabis users perceived lower selfefficacy to not use cannabis for all items assessed, such as when attending a party or celebration and General mean Mean nonusers Mean users F pN = 364 N = 294 N = 70 My father 0.28 0.27 0.33 0.100 0.753 My sisters 1.84 1.99 1.24 2.854 0.092 My brothers 1.73 1.73 1.70 0.006 0.936 My friends (female) 0.21 0.18 0.33 3.791 0.052 My friends (male) 0.35 0.23 0.86 41.571 0.000 My best friend 0.34 0.26 0.69 6.885 0.009 My boyfriend/girlfriend 2.03 2.08 1.81 0.340 0.560 My classmates (female) 0.16 0.10 0.39 9.950 0.002 My classmates (male) 0.13 0.10 0.27 8.069 0.005 Selfefficacy (m 5)d How difficult is it for you not to use cannabis in the following situations? I feel great 3.45 3.81 2.70 117.766 0.000 I'm at a party or celebration 3.30 3.56 2.17 114.085 0.000 I'm worried 3.46 3.68 2.53 90.718 0.000 I'm stressed 3.38 3.62 2.40 82.108 0.000 In others' houses or mine when the parents are NOT 3.67 3.84 2.96 79.736 0.000 My friends (male) 3.53 3.71 2.79 66.215 0.000 In public spaces 3.79 2.96 2.96 64.565 0.000 When I'm alone 3.70 3.84 3.09 58.686 0.000 I'm with others who consume 3.49 3.66 2.76 53.769 0.000 My friends (female) 3.58 3.72 2.99 44.119 0.000 Upon returning home after class 3.83 3.84 3.54 20.761 0.000 Intention (m 5)e Do you intend to use cannabis in the future? 0.48 0.28 1.33 101.372 0.000 Do you intend to use cannabis next year? 0.40 0.17 1.36 160.897 0.000 Abbreviation: (m X), number of missing values per variable. aAnswer coding for advantages and disadvantages 1 = Strongly disagree; 5 = Strongly agree. bAnswer coding for social norm: 1 = I definitely should not use cannabis; 5 = I definitely should use cannabis. cAnswer coding for social modeling & pressure: 0 = never, 4 = always. dAnswer coding for selfefficacy: 0 = Very difficult not use cannabis, 4 = Very easy not use cannabis. eAnswer coding for intention: Surely no = 0, Surely yes = 4. TABLE 1 (Continued) 15475069, 2023, 3, Downloaded from https://sigmapubs.onlinelibrary.wiley.com/doi/10.1111/jnu.12806 by Universidad De Sevilla, Wiley Online Library on [06/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 745 CANNABIS USE FACTORS IN ADOLESCENTS when feeling stressed or worried. Moreover, feeling great was a risk situation that reduced their selfefficacy. Factors uniquely associated with cannabis use in the last 12 months To determine the factors uniquely associated with cannabis use in the last 12 months, a model was developed using logistic regression analysis. The results revealed that gender and age were only significant in model 1. Models 2 and 3 revealed similar outcomes, showing increased risks of using cannabis in adolescents with a dysfunctional family, a low family affluence, receiving weekly pocket money of at least €20, having a positive attitude toward cannabis, perceiving individuals using cannabis, social norms from others favoring cannabis, feeling pressure to use cannabis, not feeling capable of saying no to consume cannabis and the intention to use cannabis in the future. Nagelkerke's indicator of explained variance was 60% (Table 4). DISCUSSION This study focused on the association between sociodemographic and motivational factors and cannabis use among a sample of grades 10– 11 and VT students in Andalusia (aged between 15 and 18 years). Our results showed the importance of these factors in cannabis use. These findings agree with those reported in other international studies (De Vries et al., 1995; Frank & Fiegel, 2020; Jalilian et al., 2020). However, they have also revealed important contextual findings about family affluence, weekly pocket money, the disadvantages and advantages of cannabis use and social influences on cannabis use. They are described below. The findings provide directions concerning important factors for Andalusia and other countries. Yet, as perceptions concerning cannabis may differ per culture and subpopulations, similar analysis per country and subgroups (e.g. differing in age and socioeconomical background) is recommended. Regarding sociodemographic factors, as age increases, cannabis use might become riskier, as already highlighted in the literature (Agency for Social Services and Dependency of Andalusia, 2018; European Monitoring Centre for Drugs and Drug Addiction, 2021; Spanish Observatory of Drugs and Addictions, 2021; World Drug Report, 2021). Moreover, there is evidence that a dysfunctional family is a predictive factor for cannabis abuse (Terzic Supic et al., 2013). When the family is dysfunctional, the adolescent may not have sufficient defense mechanisms to avoid cannabis use (i.e. coping responses, use of resources and decision making). Focusing on socioeconomic status, a study by Ter Bogt involving 30 countries has revealed that low family affluence is associated with cannabis use, which agrees with our results (ter Bogt et al., 2014). Furthermore, this study replicated earlier Spanish and international findings where a greater availability of money by the adolescent is positively related to greater consumption (Burdzovic Andreas et al., 2016; Rial et al., 2018). A recent study has simultaneously evaluated family affluence and pocket money in Spain, which showed results similar to those reported in this study (DíazGeada et al., 2021). Nevertheless, continuing studying this finding would be interesting to assess whether it is replicated with other samples, in longitudinal studies, or through a deep analysis of the relationship of cannabis use with other explicative variables. Cannabis users were less convinced that cannabis use is bad for their health, causes family problems and generates regret than nonusers. Moreover, a French study has shown that users did not clearly know the disadvantages of cannabis (Chabrol et al., 2004). A recent Spanish study (BelzuneguiEraso et al., 2020) has also concluded that information provided to individuals about cannabis is not related to health consequences (e.g. socio– cognitive and physical effect, treatment and poisonings) but to useful therapeutic effects, ways of using, or its legalization (AlonsoAyuso, 2020). Consequently, Spanish campaigns may not have had the expected results. Therefore, more information about detrimental effects is needed in prevention campaigns because users tended to downplay these effects and to optimally inform nonusers about the detrimental effects of using cannabis. Concerning the advantages of cannabis use, users were more convinced of all advantages than nonusers, most notably, concerning relaxation and medical use. Moreover, several studies (BelzuneguiEraso et al., 2020; Chabrol et al., 2004; Johnson et al., 2021) have underlined that cannabis users highlight relaxation as the main pleasant effect of cannabis use. These consequences were also regarded by nonusers as an advantage. As cannabis use has become more accepted for recreational use in Spain (AlonsoAyuso, 2020; BelzuneguiEraso et al., 2020), this TABLE 2 Correlations between cannabis use in the last 12 months and sociodemographic and motivational factors Cannabis use last 12 months (Pearson correlation) p Age 0.166a0.001 Gender −0.139a0.008 Grade 0.111 0.035 Mother educational level 0.034 0.526 Father educational level −0.003 0.951 Boyfriend/girlfriend 0.090 0.087 Nationality 0.057 0.281 Religion −0.030 0.572 Weekly pocket money −0.075 0.155 Family affluence 0.122 0.022 Family function 0.284a0.000 Attitude cons −0.497a0.000 Attitude pros 0.286a0.000 Social modeling 0.460a0.000 Social norm 0.368a0.000 Social pressure 0.221a0.000 Selfefficacy −0.500a0.000 Intention 0.532a0.000 aCorrelation is significant at the 0.01 level. 15475069, 2023, 3, Downloaded from https://sigmapubs.onlinelibrary.wiley.com/doi/10.1111/jnu.12806 by Universidad De Sevilla, Wiley Online Library on [06/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 746 | CANNABIS USE FACTORS IN ADOLESCENTS TABLE 3 Baseline characteristics of participants by cannabis consumption: nonusers versus users Baseline Variable Total Nonusers Users Test Statistic pN = 364 N = 294 N = 70 Cannabis use: last 12 months (m 5) Never 294 (78.4%) 1– 3 day 38 (10.3%) 4– 9 days 11 (3%) 10– 19 days 5 (1.4%) 20 days or more 16 (4.3%) Age (15– 18) (mean, SD) (m 0) 16.42 (2.38, 1.068) 16.34 (2.29, 1.042) 16.96 (2.79, 1.089) F (12.799) 0.000 Gender (m 0) Male 172 (46.6%) 129 (43%) 43 (61.4%) χ2 (6.987) 0.006 Female 197 (53.4%) 165 (45.32%) 27 (38.6%) City (m 0) Cádiz 152 (40.5%) 118 (40.1%) 33 (47.1%) χ2 (1.145) 0.564 Córdoba 135 (36%) 110 (37.4%) 23 (32.9%) Sevilla 82 (21.9%) 66 (22.4%) 14 (20%) Adolescents educational level (m 0) 10th grade 207 (56.1%) 174 (59.2%) 33 (47.1%) 11th grade 114 (30.9%) 88 (29.9%) 24 (34.3%) χ2 (4.492) 0.106 VT 48 (13%) 32 (10.9%) 13 (18.6%) Mother educational level (m 4) None 25 (6.8%) 19 (6.5%) 6 (8.6%) 10th grade 171 (46.3%) 139 (47.3%) 31 (44.3%) χ2 (2.043) 0.843 11th grade/VT 97 (26.3%) 78 (26.5%) 18 (25.7%) University studies 41 (11.1%) 33 (11.22%) 8 (11.4%) I do not know or not have 31 (8.4%) 21 (7.14%) 7 (10%) Father educational level (m 3) None 37 (10%) 25 (8.5%) 12 (17.4%) 10th grade 164 (44.4%) 139 (47.3%) 25 (36.2%) χ2 (6.275) 0.280 11th grade/VT 81 (22%) 64 (21.8%) 16 (23.2%) University studies 40 (10.8%) 32 (10.9%) 8 (11.6%) I do not know or do not have 44 (11.9%) 32 (10.9%) 8 (11.6%) Boyfriend/girlfriend (m 4) No 231 (62.6%) 190 (64.6%) 38 (54.3%) χ2 (3.676) 0.159 Yes 134 (36.3%) 101 (34.4%) 32 (45.7%) Nationality (m 2) Spanish 363 (98.4%) 290 (98.6%) 68 (97.1%) χ2 (6.146) 0.188 NonSpanish 4 (1.1%) 2 (0.7%) 2 (2.9%) Religion (m 3) Catholic 255 (69.1%) 206 (70.1%) 44 (62.9%) χ2 (8.49) 0.131 Others 10 (2.7%) 8 (2.72%) 2 (2.9%) No religion 101 (27.4%) 77 (26.2%) 24 (34.3%) Apgar test (m 0) Dysfunctional family 82 (22.2%) 50 (16.7%) 32 (45.7%) χ2 (29.404) 0.000 Functional family 287 (77.8%) 249 (83.7%) 38 (54.3%) 15475069, 2023, 3, Downloaded from https://sigmapubs.onlinelibrary.wiley.com/doi/10.1111/jnu.12806 by Universidad De Sevilla, Wiley Online Library on [06/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 747 CANNABIS USE FACTORS IN ADOLESCENTS may have contributed to perceiving advantages also in the nonuser group. It may be that the current debates about legalizing marijuana may have led to such beliefs. Furthermore, studies (European Monitoring Centre for Drugs and Drug Addiction, 2021; Johnson et al., 2021) have reported that positive perceptions toward the medicinal use of cannabis may increase its recreational use. Consequently, more research is needed on how to best balance information about the medical use of cannabis without increasing the risk of using cannabis recreationally. Additionally, most studies have identified the importance of social influences on cannabis use, particularly by peers (Burdzovic Andreas et al., 2016; Chabrol et al., 2004; O'Loughlin et al., 2019). We found similar findings with more social influences favoring cannabis use by friends, although findings differed slightly for social norms, modeling and pressure. For instance, a Norwegian study has found an increased likelihood of cannabis use when having romantic partners who were cannabis users (Burdzovic Andreas et al., 2016). Other studies (BelzuneguiEraso et al., 2020; O'Loughlin et al., 2019) have suggested that parental cannabis use enables adolescent's use. Our participants' parents appear not to use cannabis, and students are certain that their parents think that they should not use cannabis. Another interesting finding was the relatively positive social influences from siblings. Some studies (O'Loughlin et al., 2019; Walker et al., 2011) have also reported similar findings, suggesting that cannabis use at home is one of the better predictors of its consumption by adolescents. The findings in this study showed a difference in cannabis use among males (61.4%) and females (38.6%). This is unlike other Baseline Variable Total Nonusers Users Test Statistic pN = 364 N = 294 N = 70 Family affluence (m 16) Low 11 (3.1%) 7 (2.45%) 4 (5.9%) Medium 126 (35.7%) 96 (33.7%) 30 (44.1%) χ2 (5.375) 0.068 High 216 (61.2%) 181 (63.9%) 34 (50%) Weekly pocket money (m 0) 0– 10 € 152 (41.2%) 122 (40.8%) 30 (42.8%) 1 1 – 2 0 € 70 (19%) 55 (18.4%) 15 (21.4%) χ2 (19.846) <0.001 +20€ 29 (8.7%) 15 (6%) 14 (20%) Abbreviations: (m X), number of missing values per variable; F, ANOVA factor; VT, vocational training; χ2, Chi square. TABLE 3 (Continued) TABLE 4 Resultant variables in logistic regression analysis for cannabis use Variables Model 1 Model 2 Model 3 pOR IC 95% OR pOR IC 95% OR pOR IC 95% OR Inf Sup Inf Sup Inf Sup Gender (female vs. male) 0.000 0.522 0.405 0.672 0.123 0.765 0.545 1.075 0.421 0.867 0.613 1.227 Age 0.000 1.226 1.135 1.324 0.057 1.101 0.997 1.216 0.014 1.131 1.026 1.247 Family function 0.000 0.752 0.717 0.788 0.000 0.843 0.792 0.897 0.000 0.863 0.811 0.919 Family affluence 0.000 0.850 0.793 0.912 0.000 0.790 0.720 0.867 0.000 0.790 0.719 0.868 Weekly pocket money 0.000 1.761 1.493 2.078 0.000 2.331 1.858 2.923 0.000 2.296 1.827 2.884 Disadvantages 0.000 0.871 0.840 0.903 0.000 0.873 0.842 0.905 Advantages 0.897 1.002 0.966 1.040 0.489 0.987 0.950 1.025 Social modeling 0.000 1.103 1.062 1.146 0.000 1.087 1.045 1.130 Social norm 0.000 1.204 1.156 1.255 0.000 1.195 1.146 1.246 Social pressure 0.000 0.912 0.869 0.958 0.000 0.912 0.867 0.958 Selfefficacy 0.000 0.916 0.898 0.935 0.000 0.937 0.916 0.959 Intention 0.000 1.291 1.150 1.449 Chisquare omnibus 294.440 901.643 924.571 Nagelkerke R20.225 0.595 0.605 Abbreviations: 95% CI, 95% confidence interval; OR, odds ratio; p, p value. 15475069, 2023, 3, Downloaded from https://sigmapubs.onlinelibrary.wiley.com/doi/10.1111/jnu.12806 by Universidad De Sevilla, Wiley Online Library on [06/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License